[gentoo-science] [sage-on-gentoo] sage-5.4.1-r3 build failure

[gentoo-science] [sage-on-gentoo] sage-5.4.1-r3 build failure

[v_2e]

[-- Attachment #1: Type: text/plain, Size: 670 bytes --]

  Hello!
  I tried to upgrade to Sage-5.4.1-r3 today, but the build process
failed with this error message:

x86_64-pc-linux-gnu-gcc -pthread -shared -Wl,-O1 -Wl,--as-needed
-march=k8 -O2 -pipe
-fno-strict-aliasing /var/tmp/portage/sci-mathematics/sage-5.4.1-r3/work/sage-5.4.1-python2_7/temp.linux-x86_64-2.7/sage/libs/pari/gen.o
-L/usr//lib -L/usr/lib64 -lcsage -lpari -lgmp -lstdc++ -lntl -lpari
-lgmp -lpython2.7
-o /var/tmp/portage/sci-mathematics/sage-5.4.1-r3/work/sage-5.4.1-python2_7/lib/sage/libs/pari/gen.so
error: command 'x86_64-pc-linux-gnu-g++' failed with exit status 1

A complete build log is attached.

  Regards,
    Vladimir

----- 
 <v_2e@ukr.net>

[-- Attachment #2: sci-mathematics:sage-5.4.1-r3:20121226-144821.log.xz --]
[-- Type: application/x-xz, Size: 32344 bytes --]

Re: [gentoo-science] [sage-on-gentoo] sage-5.4.1-r3 build failure

[Francois Bissey]

On 27/12/12 05:27, v_2e@ukr.net wrote:
>   Hello!
>   I tried to upgrade to Sage-5.4.1-r3 today, but the build process
> failed with this error message:
> 
> x86_64-pc-linux-gnu-gcc -pthread -shared -Wl,-O1 -Wl,--as-needed
> -march=k8 -O2 -pipe
> -fno-strict-aliasing /var/tmp/portage/sci-mathematics/sage-5.4.1-r3/work/sage-5.4.1-python2_7/temp.linux-x86_64-2.7/sage/libs/pari/gen.o
> -L/usr//lib -L/usr/lib64 -lcsage -lpari -lgmp -lstdc++ -lntl -lpari
> -lgmp -lpython2.7
> -o /var/tmp/portage/sci-mathematics/sage-5.4.1-r3/work/sage-5.4.1-python2_7/lib/sage/libs/pari/gen.so
> error: command 'x86_64-pc-linux-gnu-g++' failed with exit status 1
> 
> A complete build log is attached.
> 

Hi Vladimir,

I am afraid I couldn't read the build log. The encoding went all wrong
apparently. I am more suspicious that the compression/decompression
didn't go right. Could you send it gzipped or bzipped just to check.

Francois

Re: [gentoo-science] [sage-on-gentoo] sage-5.4.1-r3 build failure

[Vladimir Smolyar]

[-- Attachment #1: Type: text/plain, Size: 743 bytes --]

  Hello!

On Thu, 27 Dec 2012 21:20:13 +1300
Francois Bissey <fbissey@slingshot.co.nz> wrote:

> On 27/12/12 05:27, v_2e@ukr.net wrote:
> > . . .
> > A complete build log is attached.
> > 
> 
> Hi Vladimir,
> 
> I am afraid I couldn't read the build log. The encoding went all wrong
> apparently. I am more suspicious that the compression/decompression
> didn't go right. Could you send it gzipped or bzipped just to check.
> 
  No problem! Here is the bzip'ped one.
  I noticed that some kind of escape-sequences for colouring purposes
are there in the log file. Maybe it is because of them that the file is
difficult to read at certain places?

  Regards,
    Vladimir

----- 
Vladimir Smolyar
E-mail: v_2e@ukr.net
Jabber: v_2e@jabber.od.ua

[-- Attachment #2: sci-mathematics:sage-5.4.1-r3:20121226-144821.log.bz2 --]
[-- Type: application/x-bzip2, Size: 28398 bytes --]

Re: [gentoo-science] [sage-on-gentoo] sage-5.4.1-r3 build failure

[Francois Bissey]

On 27/12/12 22:03, Vladimir Smolyar wrote:
>   Hello!
> 
> On Thu, 27 Dec 2012 21:20:13 +1300
> Francois Bissey <fbissey@slingshot.co.nz> wrote:
> 
>> On 27/12/12 05:27, v_2e@ukr.net wrote:
>>> . . .
>>> A complete build log is attached.
>>>
>>
>> Hi Vladimir,
>>
>> I am afraid I couldn't read the build log. The encoding went all wrong
>> apparently. I am more suspicious that the compression/decompression
>> didn't go right. Could you send it gzipped or bzipped just to check.
>>
>   No problem! Here is the bzip'ped one.
>   I noticed that some kind of escape-sequences for colouring purposes
> are there in the log file. Maybe it is because of them that the file is
> difficult to read at certain places?

Hi,

escape sequences make things slightly more difficult to read but that
wasn't the problem. I actually only had garbage symbols. The bziped one
is much better. My decompressor may have a flaw.

Francois

Re: [gentoo-science] [sage-on-gentoo] sage-5.4.1-r3 build failure

[Francois Bissey]

On 27/12/12 22:03, Vladimir Smolyar wrote:
>   Hello!
> 
> On Thu, 27 Dec 2012 21:20:13 +1300
> Francois Bissey <fbissey@slingshot.co.nz> wrote:
> 
>> On 27/12/12 05:27, v_2e@ukr.net wrote:
>>> . . .
>>> A complete build log is attached.
>>>
>>
>> Hi Vladimir,
>>
>> I am afraid I couldn't read the build log. The encoding went all wrong
>> apparently. I am more suspicious that the compression/decompression
>> didn't go right. Could you send it gzipped or bzipped just to check.
>>
>   No problem! Here is the bzip'ped one.
>   I noticed that some kind of escape-sequences for colouring purposes
> are there in the log file. Maybe it is because of them that the file is
> difficult to read at certain places?
> 
OK got the error:
x86_64-pc-linux-gnu-g++ -pthread -march=k8 -O2 -pipe
-fno-strict-aliasing -fPIC -I/usr//include -I/usr/include/csage
-Isage/ext -I/usr/include/python2.7 -c sage/libs/ppl.cpp -o
/var/tmp/portage/sci-mathematics/sage-5.4.1-r3/work/sage-5.4.1-python2_7/temp.linux-x86_64-2.7/sage/libs/ppl.o
In file included from sage/libs/ppl.cpp:264:0:
/usr/include/ppl.hh:783:7: error: redefinition of ‘class
std::numeric_limits<__gmp_expr<__mpz_struct [1], __mpz_struct [1]> >’
/usr/include/gmpxx.h:3269:21: error: previous definition of ‘class
std::numeric_limits<__gmp_expr<__mpz_struct [1], __mpz_struct [1]> >’
/usr/include/ppl.hh:844:7: error: redefinition of ‘class
std::numeric_limits<__gmp_expr<__mpq_struct [1], __mpq_struct [1]> >’
/usr/include/gmpxx.h:3306:21: error: previous definition of ‘class
std::numeric_limits<__gmp_expr<__mpq_struct [1], __mpq_struct [1]> >’

Would I be correct in assuming you are using gcc-4.7.*? I hate c++
but it wouldn't hurt to rebuild gmp and then ppl. Which version of ppl
do you have by the way, as it could have an impact too.

Francois

Re: [gentoo-science] [sage-on-gentoo] sage-5.4.1-r3 build failure

[v_2e]

  Hello!

On Thu, 27 Dec 2012 22:54:18 +1300
Francois Bissey <fbissey@slingshot.co.nz> wrote:
> 
> Would I be correct in assuming you are using gcc-4.7.*? I hate c++
> but it wouldn't hurt to rebuild gmp and then ppl. Which version of ppl
> do you have by the way, as it could have an impact too.
> 
  No, I have only sys-devel/gcc-4.6.3 installed.
  As for the PPL, its version is dev-libs/ppl-0.12.1

  Regards, 
    Vladimir

----- 
 <v_2e@ukr.net>

Re: [gentoo-science] [sage-on-gentoo] sage-5.4.1-r3 build failure

[v_2e]

  Hello again!

On Thu, 27 Dec 2012 22:54:18 +1300
Francois Bissey <fbissey@slingshot.co.nz> wrote:

> I hate c++ but it wouldn't hurt to rebuild gmp and then ppl. 

  Unfortunately, I cannot build dev-libs/ppl-0.12.1 because of the bug
https://bugs.gentoo.org/show_bug.cgi?id=447928

  Regards,
    Vladimir

----- 
 <v_2e@ukr.net>

Re: [gentoo-science] [sage-on-gentoo] sage-5.4.1-r3 build failure

[Francois Bissey]

On 28/12/12 12:25, v_2e@ukr.net wrote:
>   Hello again!
> 
> On Thu, 27 Dec 2012 22:54:18 +1300
> Francois Bissey <fbissey@slingshot.co.nz> wrote:
> 
>> I hate c++ but it wouldn't hurt to rebuild gmp and then ppl. 
> 
>   Unfortunately, I cannot build dev-libs/ppl-0.12.1 because of the bug
> https://bugs.gentoo.org/show_bug.cgi?id=447928
> 

Hi Vladimir,

it looks like the error you get building sage is exactly the same....
Or strikingly similar. So I am expecting it to be a manifestation of the
same bug. The patch for compiling ppl should help compiling
sage once ppl has been rebuilt with it. It looks like a serious
problem hitting a number of poeple judging by the number of duplicates.
I will give the dev a couple of days to get a solution before getting
something. Ideally we need a newer ppl from upstream if I understand the
problem correctly.

Is it urgent?

Francois

Re: [gentoo-science] [sage-on-gentoo] sage-5.4.1-r3 build failure

[v_2e]

  Hello!

On Fri, 28 Dec 2012 20:47:31 +1300
Francois Bissey <fbissey@slingshot.co.nz> wrote:
> 
> Hi Vladimir,
> 
> it looks like the error you get building sage is exactly the same....
> Or strikingly similar. So I am expecting it to be a manifestation of
> the same bug. The patch for compiling ppl should help compiling
> sage once ppl has been rebuilt with it. It looks like a serious
> problem hitting a number of poeple judging by the number of
> duplicates. I will give the dev a couple of days to get a solution
> before getting something. Ideally we need a newer ppl from upstream
> if I understand the problem correctly.
> 
> Is it urgent?
> 
> Francois
> 
  I managed to build Sage-5.5 using the patch uploaded here:
https://bugs.gentoo.org/show_bug.cgi?id=447928
  I did not test it for Sage-5.4, but I don't think it really matters
now when Sage-5.5 is out, or does it?

  Regards,
    Vladimir

-----
v_2e <v_2e@ukr.net>

Re: [gentoo-science] [sage-on-gentoo] sage-5.4.1-r3 build failure

[Francois Bissey]

On 31/12/12 07:52, v_2e wrote:
>   Hello!
> 
> On Fri, 28 Dec 2012 20:47:31 +1300
> Francois Bissey <fbissey@slingshot.co.nz> wrote:
>>
>> Hi Vladimir,
>>
>> it looks like the error you get building sage is exactly the same....
>> Or strikingly similar. So I am expecting it to be a manifestation of
>> the same bug. The patch for compiling ppl should help compiling
>> sage once ppl has been rebuilt with it. It looks like a serious
>> problem hitting a number of poeple judging by the number of
>> duplicates. I will give the dev a couple of days to get a solution
>> before getting something. Ideally we need a newer ppl from upstream
>> if I understand the problem correctly.
>>
>> Is it urgent?
>>
>> Francois
>>
>   I managed to build Sage-5.5 using the patch uploaded here:
> https://bugs.gentoo.org/show_bug.cgi?id=447928
>   I did not test it for Sage-5.4, but I don't think it really matters
> now when Sage-5.5 is out, or does it?
> 

Hi Vladimir,

it indeed doesn't matter about 5.4 but it would be exactly the same.

Francois